Molecular and morphometric description of adipose tissue during weight changes: a quantitative tool for assessment of tissue texture.

The aim of this study was to evaluate the morphometric changes of adipose tissue of lean and obese rats as assessed by computerized image analysis (IA) system in experimental conditions, with different degrees of adiposity. Moreover, to validate measures obtained by image analysis by correlation with direct measures of adiposity (body weight, epididimal fat, mean fat cell size and serum leptin). Finally to correlate these changes to expression of genes involved in lipid deposition and mobilization in adipose tissue. Lean (Fa/?) and genetically obese (fa/fa) Zucker rats were studied. Obese rats were food-restricted or treated with retinoic acid (ATRA) in order to reduce body weight and fat content. Moreover, gene expression of two key enzymes involved in fat metabolism (HSL and DGAT) were assessed in adipose tissue by RT-PCR. Our results show that HSL expression in adipose tissue was lower in obese compared to lean rats (1.47+/-0.02 vs 0.35+/-0.03, p<0.005) and was upregulated during food restriction in obese rats. DGAT expression was similar in lean and obese rats and was reduced by treatment with ATRA in obese rats. Tissue texture assessed by IA was significantly higher in lean compared to obese rats (23.2+/-0.6 vs 11.6+/-2.4%; p=0.01). Tissue structure highly correlated with adiposity in obese rats with different amount of body fat (area fraction vs epididimal fat depot: p=0.001). Distribution of measures for each sample, an index of spread of adipose tissue texture, as expressed by the standard deviation, correlated with adiposity (standard deviation vs epididimal fat depot: p=0.002) thus suggesting that adipose tissue texture increases its heterogeneity when adiposity is lower. This observation is in agreement with the hypothesis that the process of lipid mobilization from adipose tissue is not uniform, but a subpopulation of slimming adipocytes undergoes a complete release of their fat content while the rest of the tissue is much less affected. Moreover, image analysis system seems a reliable quantitative tool for assessment of adipose tissue texture.